1900.] on High-Speed Navigation Steam Turbines. 243 



above the propeller by an air pump. Glass windows are fitted for ob- 

 servation and illumination. Under these conditions the only forces 

 tending to hold the water together and resist cavitation are the small 

 head of water above the propeller, and capillarity. The propeller is 

 2 inches diameter and 3 inches pitch ; cavitation commences at about 

 1200 revolutions and becomes very pronounced at 1500 revolutions. 

 Had the atmospheric pressure not been removed, speeds of 12,000 and 

 15,0u0 revolutions per minute would have been necessary, rendering 

 observations more difficult. 



The arrangement we have now was kindly suggested by Mr. Heath, 

 and is a decided improvement, the revolving disc with narrow slots 

 synchronising approximately with the revolutions of the propeller. 

 The propeller is now seen to rotate very slowly, it also permits of the 

 projection of the phenomenon on the screen, which was not possible 

 with my previous arrangement. The permanence of the vortices be- 

 hind the blades is very striking. The inference to be drawn from 

 these experiments seems to be that for fast speeds of vessels, wide thin 

 blades, a coarse pitch ratio, and moderate slip, are desirable for the 

 prevention of cavitation, and in order to obtain the best efficiency in 

 propulsion of the vessel. 



To return to the Turbinia, a radical alteration was deemed neces- 

 sary. A new turbine engine was made, consisting of three separate 

 engines, high pressure, intermediate pressure, and low pressure, each 

 of which drove one screw shaft, the power of the engine was dis- 

 tributed over three shafts instead of concentrated on one, and 

 three propellers were placed on each shaft. The result of these 

 chauges was marvellous. The vessel now nearly doubled her speed, 

 30 knots was soon reached, and finally 32| knots mean speed on the 

 measured mile authenticated, or the fastest speed then attained by any 

 vessel afloat. The economy of her engines was investigated by Pro- 

 fessor Ewing, assisted by Professor Dunkerly : the consumption of 

 steam per indicated horse-power for all purposes at 31 knots speed 

 was found to be 14^ lbs., or in other words, with a good marine boiler 

 the coal consumption would be considerably under 2 lbs. per indicated 

 horse-power, a result better than is obtained in toz'pedo-boats or 

 torpedo-boat destroyers with ordinary triple expansion engines. 



The vessel's reversing turbine gave her an astern speed of 6.^ knots, 

 and she could be brought to rest in 36 seconds when running at 30 

 knots speed, and from rest she could be brought up to 30 knots in 

 40 seconds. 



The Turbinia cruised from the Tyne to the Naval Eeview at 

 Spithead, where she steamed on the day of the Review at an estimated 

 speed of 34^ knots. These results represent about 2300 indicated 

 horse-power, and may be said to have been obtained without a very 

 abnormal performance as regards the boiler ; its total heating surface 

 being 1100 square feet, and an evaporation of about 28 lbs. per 

 square foot at the speed of 34^ knots. 



These speeds were not obtained by bottling up the steam and 



